Virtually all aspects of cell growth and metabolism depend upon the precise control of protein levels. This is achieved not only by the regulation of synthesis, but also through mechanisms of protein turnover. In eukaryotes, a major pathway of intracellular proteolysis involves the attachment of the protein ubiquitin (Ub) to a protein substrate. Often, the addition of multiple Ub moieties forms a 'polyubiquitin: chain on the substrate. This structure facilitates substrate recognition and degradation by the ATP-dependent 26S protease complex. The specificity of Ub-dependent degradation had appeared to depend only on the initial selection of protein targets, but new evidence suggests that there is a second level of specificity in which ub-protein conjugates are partitioned among two alternative fates: (i) degradation of the protein substrate by the 26S protease, or (ii) deubiquitination by enzymes called isopeptidases to regenerate the intact protein substrate. An isopeptidase found in PA700, a 19S subcomplex of the 26S protease, is a likely candidate for this 'editing' activity. Experiments will be done to investigate the selection and degradation of (poly) Ub-protein conjugates by the 26S protease, and to determine the role of the PA700-isopeptidase in this process. For this purpose, uniform polyUb chains and polyUb-protein conjugates will be prepared for use as substrates and ligands with purified bovine PA700 and 26S complexes. Rates and products of 26S-catalyzed conjugate degradation will by analyzed as a function of polyUb chainlength and protein substrate structure. Methods will be developed to direct the Ub-Ub linkages in polyUb chains through any Ub lysine residue. Binding affinities and numbers of sites on PA700 and 26S particles by electron microscopy, gold-labeled proteins that bind specifically to these sites will be developed. In vivo functions for the PA700-isopeptidase will be explored by genetic and biochemical studies in yeast. Sequence from the bovine isopeptidase will be used to identify and clone its equivalent in Saccharomyces cerevisiae. Mutagenesis will be done to inactivative the enzyme, and phenotypes that might be expected from loss of editing functions will be tested.